BIOS upgrades - reflashing the BIOS

[Bill in Co.]

Just spent an "interesting" (to put it mildly!) weekend on this project to
enable a microprocessor upgrade. I wouldn't recommend it for the
faint-hearted, suffice it to say, or you might end up with a paperweight.


You know, it wouldn't be quite so bad, except that in some cases, there is
no reverse path available (i..e, it's a one way trip) - which seems
unforgiveable, on their part!

My question is: why do these BIOS manufacturers sometimes make it so damn
difficult to upgrade/downgrade or go back to a previous version (especially
when going from one manufacturer to one its subsidiaries, or vice versa
(like Dell and Intel)?

And not only that, but even block certain upgrades with certain chips, just
out of self-centered arrogance on their part (and not due to real chip
limitations)?

 

[MEB]

"Bill in Co." <not_really_here@earthlink.net> wrote in message
news:uVIv54$gIHA.1212@TK2MSFTNGP05.phx.gbl...
| Just spent an "interesting" (to put it mildly!) weekend on this project to
| enable a microprocessor upgrade. I wouldn't recommend it for the
| faint-hearted, suffice it to say, or you might end up with a paperweight.
|
|
| You know, it wouldn't be quite so bad, except that in some cases, there is
| no reverse path available (i..e, it's a one way trip) - which seems
| unforgiveable, on their part!
|
| My question is: why do these BIOS manufacturers sometimes make it so damn
| difficult to upgrade/downgrade or go back to a previous version
(especially
| when going from one manufacturer to one its subsidiaries, or vice versa
| (like Dell and Intel)?
|
| And not only that, but even block certain upgrades with certain chips,
just
| out of self-centered arrogance on their part (and not due to real chip
| limitations)?

I can definitely relate to that type of experience. Made a few
paper-weights in the early days, a couple my fault [like flashing with the
wrong BIOS or leaving the flash disk in the wrong computer and rebooting],
some the flash tool or a badly created upgrade.

Generally, the preliminaries are always to set the BIOS back to defaults
prior to any flash [unless you're over-clocking].
Regarding the flash, if not BIOS instituted {newer boards}, the separate
flash tool/file usually has a switch to save the present BIOS to a bin or
other file. Sadly many make that information hard to find.
If this is done, then one can roll-back to the older IF you also over-write
the Boot code.

--

MEB
http://peoplescounsel.orgfree.com
_________

 

[Bill in Co.]

MEB wrote:
> "Bill in Co." <not_really_here@earthlink.net> wrote in message
> news:uVIv54$gIHA.1212@TK2MSFTNGP05.phx.gbl...
>> Just spent an "interesting" (to put it mildly!) weekend on this project
>> to
>> enable a microprocessor upgrade. I wouldn't recommend it for the
>> faint-hearted, suffice it to say, or you might end up with a paperweight.
>>
>>
>> You know, it wouldn't be quite so bad, except that in some cases, there
>> is
>> no reverse path available (i..e, it's a one way trip) - which seems
>> unforgiveable, on their part!
>>
>> My question is: why do these BIOS manufacturers sometimes make it so
>> damn
>> difficult to upgrade/downgrade or go back to a previous version
>> (especially
>> when going from one manufacturer to one its subsidiaries, or vice versa
>> (like Dell and Intel)?
>>
>> And not only that, but even block certain upgrades with certain chips,
>> just
>> out of self-centered arrogance on their part (and not due to real chip
>> limitations)?
>
> I can definitely relate to that type of experience. Made a few
> paper-weights in the early days, a couple my fault [like flashing with the
> wrong BIOS or leaving the flash disk in the wrong computer and rebooting],
> some the flash tool or a badly created upgrade.

I bet that was *quite* annoying, too.

> Generally, the preliminaries are always to set the BIOS back to defaults
> prior to any flash [unless you're over-clocking].

Also, sometimes it appears that *after* a flash is done, one may have to
first load its defaults BEFORE trying to reconfigure it for your system.
Learned that the hard way, but at least I didn't end up with a paperweight
(but it was a close call, at some points).

Actually, at a later point in time, it wouldn't boot into Windows, so I had
created a DOS-only machine at that point. (Nice, really nice).
Still, it was better than being locked out of even being able to get in and
set up the BIOS for the system (it demanded a password at that point).

Finally got this all resolved. (Sigh of relief). What a weekend,
though. It can be a bit hair-raising, to say the least, since you really
are treading on the razor's edge in this territory. .

> Regarding the flash, if not BIOS instituted {newer boards}, the separate
> flash tool/file usually has a switch to save the present BIOS to a bin or
> other file. Sadly many make that information hard to find.
> If this is done, then one can roll-back to the older IF you also
> over-write
> the Boot code.

Never found that specific documentation, but I sure looked around a bit. I
don't think its an option for all of them, though. Of course, I am dealing
with an older board here, circa 2000.

 

[MEB]

"Bill in Co." <not_really_here@earthlink.net> wrote in message
news:O8SIQeBhIHA.1188@TK2MSFTNGP04.phx.gbl...
| MEB wrote:
| > "Bill in Co." <not_really_here@earthlink.net> wrote in message
| > news:uVIv54$gIHA.1212@TK2MSFTNGP05.phx.gbl...
| >> Just spent an "interesting" (to put it mildly!) weekend on this project
| >> to
| >> enable a microprocessor upgrade. I wouldn't recommend it for the
| >> faint-hearted, suffice it to say, or you might end up with a
paperweight.
| >>
| >>
| >> You know, it wouldn't be quite so bad, except that in some cases, there
| >> is
| >> no reverse path available (i..e, it's a one way trip) - which seems
| >> unforgiveable, on their part!
| >>
| >> My question is: why do these BIOS manufacturers sometimes make it so
| >> damn
| >> difficult to upgrade/downgrade or go back to a previous version
| >> (especially
| >> when going from one manufacturer to one its subsidiaries, or vice versa
| >> (like Dell and Intel)?
| >>
| >> And not only that, but even block certain upgrades with certain chips,
| >> just
| >> out of self-centered arrogance on their part (and not due to real chip
| >> limitations)?
| >
| > I can definitely relate to that type of experience. Made a few
| > paper-weights in the early days, a couple my fault [like flashing with
the
| > wrong BIOS or leaving the flash disk in the wrong computer and
rebooting],
| > some the flash tool or a badly created upgrade.
|
| I bet that was *quite* annoying, too.

More than annoying, my mistakes were costly, but getting the manufacturers
to admit that THEY had screwed up was a study in $technical war$.

|
| > Generally, the preliminaries are always to set the BIOS back to defaults
| > prior to any flash [unless you're over-clocking].
|
| Also, sometimes it appears that *after* a flash is done, one may have to
| first load its defaults BEFORE trying to reconfigure it for your system.
| Learned that the hard way, but at least I didn't end up with a paperweight
| (but it was a close call, at some points).

Right, don't set ANYTHING after that first reboot, but SAVE the CMOS
settings, reboot, then carefully walk through the prior specialty
settings.ONE AT A TIME. And this because, these may also have been modified
OR changed.

|
| Actually, at a later point in time, it wouldn't boot into Windows, so I
had
| created a DOS-only machine at that point. (Nice, really nice).
| Still, it was better than being locked out of even being able to get in
and
| set up the BIOS for the system (it demanded a password at that point).

Oh now that's interesting, a defaulted PASSWORD,,, so what BIOS/motherboard
was this?

|
| Finally got this all resolved. (Sigh of relief). What a weekend,
| though. It can be a bit hair-raising, to say the least, since you really
| are treading on the razor's edge in this territory. .
|
| > Regarding the flash, if not BIOS instituted {newer boards}, the separate
| > flash tool/file usually has a switch to save the present BIOS to a bin
or
| > other file. Sadly many make that information hard to find.
| > If this is done, then one can roll-back to the older IF you also
| > over-write
| > the Boot code.
|
| Never found that specific documentation, but I sure looked around a bit.
I
| don't think its an option for all of them, though. Of course, I am
dealing
| with an older board here, circa 2000.
|

Sounds like an old Phoenix or AMI BIOS.

--

MEB
http://peoplescounsel.orgfree.com
_________

 

[Franc Zabkar]

On Wed, 12 Mar 2008 02:03:27 -0600, "Bill in Co."
<not_really_here@earthlink.net> put finger to keyboard and composed:

>Also, sometimes it appears that *after* a flash is done, one may have to
>first load its defaults BEFORE trying to reconfigure it for your system.

The "extended" CMOS RAM (byte 65 and above) contains various chipset
register settings, eg memory timings, FSB settings, peripheral
configuration info, etc. Any RAM location can have any function as
defined by the BIOS code. For example, there is no guarantee that BIOS
version 2 will store the SDRAM memory timings at the same CMOS RAM
location as BIOS version 1, so one could expect a black screen during
the first POST after a BIOS flash. Clearing the CMOS RAM and reverting
to defaults circumvents this potential problem.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Bill in Co.]

MEB wrote:
> "Bill in Co." <not_really_here@earthlink.net> wrote in message
> news:O8SIQeBhIHA.1188@TK2MSFTNGP04.phx.gbl...
>> MEB wrote:
>>> "Bill in Co." <not_really_here@earthlink.net> wrote in message
>>> news:uVIv54$gIHA.1212@TK2MSFTNGP05.phx.gbl...
>>>> Just spent an "interesting" (to put it mildly!) weekend on this project
>>>> to
>>>> enable a microprocessor upgrade. I wouldn't recommend it for the
>>>> faint-hearted, suffice it to say, or you might end up with a
>>>> paperweight.
>>>>
>>>>
>>>> You know, it wouldn't be quite so bad, except that in some cases, there
>>>> is
>>>> no reverse path available (i..e, it's a one way trip) - which seems
>>>> unforgiveable, on their part!
>>>>
>>>> My question is: why do these BIOS manufacturers sometimes make it so
>>>> damn
>>>> difficult to upgrade/downgrade or go back to a previous version
>>>> (especially
>>>> when going from one manufacturer to one its subsidiaries, or vice versa
>>>> (like Dell and Intel)?
>>>>
>>>> And not only that, but even block certain upgrades with certain chips,
>>>> just out of self-centered arrogance on their part (and not due to real
>>>> chip
>>>> limitations)?
>>>
>>> I can definitely relate to that type of experience. Made a few
>>> paper-weights in the early days, a couple my fault [like flashing with
>>> the
>>> wrong BIOS or leaving the flash disk in the wrong computer and
>>> rebooting],
>>> some the flash tool or a badly created upgrade.
>>
>> I bet that was *quite* annoying, too.
>
> More than annoying, my mistakes were costly, but getting the manufacturers
> to admit that THEY had screwed up was a study in $technical war$.
>
>>
>>> Generally, the preliminaries are always to set the BIOS back to defaults
>>> prior to any flash [unless you're over-clocking].
>>
>> Also, sometimes it appears that *after* a flash is done, one may have to
>> first load its defaults BEFORE trying to reconfigure it for your system.
>> Learned that the hard way, but at least I didn't end up with a
>> paperweight
>> (but it was a close call, at some points).
>
> Right, don't set ANYTHING after that first reboot, but SAVE the CMOS
> settings, reboot, then carefully walk through the prior specialty
> settings.ONE AT A TIME. And this because, these may also have been
> modified
> OR changed.
>
>>
>> Actually, at a later point in time, it wouldn't boot into Windows, so I
>> had
>> created a DOS-only machine at that point. (Nice, really nice).
>> Still, it was better than being locked out of even being able to get in
>> and
>> set up the BIOS for the system (it demanded a password at that point).
>
> Oh now that's interesting, a defaulted PASSWORD,,, so what
> BIOS/motherboard
> was this?

(This is for my Dell 4100 computer, with Windows98SE (circa year 2000 or
so):

The BIOS upgrade was Intel P10 (actually EA81510A.86A.0046.P10), and the
motherboard is (Dell) Intel D815EEA.

But the problem was apparently created because when I first flashed the BIOS
and then booted up, I did NOT initially select the option to load the
Defaults (but I did try changing some of the settings in there anyway).
Big mistake. After I rebooted, and tried get into BIOS to fix things up, it
asked me for a password (real nice, I'll tell ya!! And I tried a few
entries including just <CR>, and nothing worked).

Finally, I found the way out of this hole by temporarily moving the BIOS
jumper on the motherboard to the recovery or maintenance mode position, so
that when I rebooted, I *was* finally able to get into BIOS, and change the
settings to the proper values, and no password was required.

Now the microprocessor upgrade (PowerLeap) works great, and the only problem
I have is that I cannot select Ultra DMA Mode 5 in the BIOS, as Mode 4 is as
high as it goes in the list. Bummer.

Whether or not that means I am really locked into Ultra Mode 4, or just that
the BIOS can't tell what is actually available for my system (and Mode 5
was, in the previous Dell BIOS), I don't know for sure. But I'm hoping I
am running in Ultra Mode 5 (despite what the BIOS setting says).

I probably should look for some utility that actually can report the truth
(and not just what the BIOS screen allows me to select there), as my system
IS capable of running in Ultra Mode 5. (And the newer P11 Intel BIOS
recognizes that, but it is incompatible with this microprocessor upgrade -
already BTDT).


>> Finally got this all resolved. (Sigh of relief). What a weekend,
>> though. It can be a bit hair-raising, to say the least, since you
>> really
>> are treading on the razor's edge in this territory. .
>>
>>> Regarding the flash, if not BIOS instituted {newer boards}, the separate
>>> flash tool/file usually has a switch to save the present BIOS to a bin
>>> or
>>> other file. Sadly many make that information hard to find.
>>> If this is done, then one can roll-back to the older IF you also
>>> over-write
>>> the Boot code.
>>
>> Never found that specific documentation, but I sure looked around a bit.
>> I
>> don't think its an option for all of them, though. Of course, I am
>> dealing
>> with an older board here, circa 2000.
>>
>
> Sounds like an old Phoenix or AMI BIOS.

per above...

 

[Bill in Co.]

Franc Zabkar wrote:
> On Wed, 12 Mar 2008 02:03:27 -0600, "Bill in Co."
> <not_really_here@earthlink.net> put finger to keyboard and composed:
>
>> Also, sometimes it appears that *after* a flash is done, one may have to
>> first load its defaults BEFORE trying to reconfigure it for your system.
>
> The "extended" CMOS RAM (byte 65 and above) contains various chipset
> register settings, eg memory timings, FSB settings, peripheral
> configuration info, etc. Any RAM location can have any function as
> defined by the BIOS code. For example, there is no guarantee that BIOS
> version 2 will store the SDRAM memory timings at the same CMOS RAM
> location as BIOS version 1, so one could expect a black screen during
> the first POST after a BIOS flash. Clearing the CMOS RAM and reverting
> to defaults circumvents this potential problem.
>
> - Franc Zabkar

Thanks for that info, Frank. I think I should take a look into it a bit
more.

And I also need to figure out how to get back to Ultra DMA Mode 5, since
that option wasn't available in this current BIOS. (Ultra DMA Mode 4 was as
high as it goes, or at least as it shows up in its config list). I'm not
sure if that means I'm stuck in Ultra Mode 4, or just that the BIOS doesn't
know any better, since I *do* know I was able to get it in the other BIOS
(i.e., the HD's support it). But this is the only BIOS version I can use,
for this specific microprocessor upgrade.

 

[Franc Zabkar]

On Wed, 12 Mar 2008 13:50:46 -0600, "Bill in Co."
<not_really_here@earthlink.net> put finger to keyboard and composed:

>I probably should look for some utility that actually can report the truth
>(and not just what the BIOS screen allows me to select there), as my system
>IS capable of running in Ultra Mode 5. (And the newer P11 Intel BIOS
>recognizes that, but it is incompatible with this microprocessor upgrade -
>already BTDT).

How about SmartUDM? It's a DOS application, so it should tell you how
the drive is configured after the POST.

Here are some example reports from my PCs:

http://www.users.on.net/~fzabkar/SmartUDM/120GB.RPT

HDD 1 Model: ST3120026A (Athlon XP 2500 CPU)

PIO Mode Support: 4
SW DMA Mode Support: None
MW DMA Mode Support: 2, Active: None
UDMA Mode Support: 5 (UltraDMA/100), Active: 5

http://www.users.on.net/~fzabkar/SmartUDM/13GB.RPT

HDD 1 Model: ST313021A (socket 7 motherboard)

PIO Mode Support: 4
SW DMA Mode Support: None
MW DMA Mode Support: 2, Active: None
UDMA Mode Support: 4 (UltraDMA/66), Active: 2

http://www.users.on.net/~fzabkar/SmartUDM/6GB.RPT

HDD 2 Model: FUJITSU MPE3064AT (Athlon XP 2500, IIRC)

PIO Mode Support: 4
SW DMA Mode Support: None
MW DMA Mode Support: 2, Active: None
UDMA Mode Support: 4 (UltraDMA/66), Active: 4

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Franc Zabkar]

On Thu, 13 Mar 2008 08:52:10 +1100, Franc Zabkar
<fzabkar@iinternode.on.net> put finger to keyboard and composed:

>On Wed, 12 Mar 2008 13:50:46 -0600, "Bill in Co."
><not_really_here@earthlink.net> put finger to keyboard and composed:
>
>>I probably should look for some utility that actually can report the truth
>>(and not just what the BIOS screen allows me to select there), as my system
>>IS capable of running in Ultra Mode 5. (And the newer P11 Intel BIOS
>>recognizes that, but it is incompatible with this microprocessor upgrade -
>>already BTDT).
>
>How about SmartUDM? It's a DOS application, so it should tell you how
>the drive is configured after the POST.

Everest Home Edition (for Windows) produces the following report for a
Fujitsu MPF3204AT in a socket 7 box:

Max. PIO Transfer Mode PIO 4
Max. UDMA Transfer Mode UDMA 4 (ATA-66)
Active UDMA Transfer Mode UDMA 2 (ATA-33)

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Bill in Co.]

Franc Zabkar wrote:
> On Thu, 13 Mar 2008 08:52:10 +1100, Franc Zabkar
> <fzabkar@iinternode.on.net> put finger to keyboard and composed:
>
>> On Wed, 12 Mar 2008 13:50:46 -0600, "Bill in Co."
>> <not_really_here@earthlink.net> put finger to keyboard and composed:
>>
>>> I probably should look for some utility that actually can report the
>>> truth
>>> (and not just what the BIOS screen allows me to select there), as my
>>> system
>>> IS capable of running in Ultra Mode 5. (And the newer P11 Intel BIOS
>>> recognizes that, but it is incompatible with this microprocessor
>>> upgrade -
>>> already BTDT).
>>
>> How about SmartUDM? It's a DOS application, so it should tell you how
>> the drive is configured after the POST.
>
> Everest Home Edition (for Windows) produces the following report for a
> Fujitsu MPF3204AT in a socket 7 box:
>
> Max. PIO Transfer Mode PIO 4
> Max. UDMA Transfer Mode UDMA 4 (ATA-66)
> Active UDMA Transfer Mode UDMA 2 (ATA-33)
>
> - Franc Zabkar

Thanks, Frank. I'll look into both of them (and I already have an older
freebie version of Everest (1.51)

Hmmm. ...let's see...
OK, found it in Everest, under the "Storage, ATA" tab, I get:

Max. PIO Transfer Mode: PIO 4
Max. UDMA Transfer Mode: UDMA 5 (ATA-100)
Active UDMA Transfer Mode: UDMA 5 (ATA-100)

....which sounds encouraging. I think it's this high because I'm using
the 80 conductor cables, and my WD EIDE drives support it, and, as I recall,
I installed some Intel software update at one point in time (but NOT that
infamous one, which trashed my system on another date (and there was one of
those, shown as a recommended update, on the MS site, but they later removed
it).

So I'm wondering why yours is a bit lower above, and I expect it's due to
some condition mentioned above not being met (or perhaps a really old BIOS)?

But at any rate, evidently this BIOS doesn't show Ultra Mode 5 (because it
is too dated), but apparently the system IS running with Ultra Mode 5
capability.

I should probably try out the DOS version to really confirm this, too.

Thanks.

 

[Franc Zabkar]

On Wed, 12 Mar 2008 21:53:31 -0600, "Bill in Co."
<not_really_here@earthlink.net> put finger to keyboard and composed:

>Franc Zabkar wrote:

>> Everest Home Edition (for Windows) produces the following report for a
>> Fujitsu MPF3204AT in a socket 7 box:
>>
>> Max. PIO Transfer Mode PIO 4
>> Max. UDMA Transfer Mode UDMA 4 (ATA-66)
>> Active UDMA Transfer Mode UDMA 2 (ATA-33)
>>
>> - Franc Zabkar
>
>Thanks, Frank. I'll look into both of them (and I already have an older
>freebie version of Everest (1.51)
>
>Hmmm. ...let's see...
>OK, found it in Everest, under the "Storage, ATA" tab, I get:
>
>Max. PIO Transfer Mode: PIO 4
>Max. UDMA Transfer Mode: UDMA 5 (ATA-100)
>Active UDMA Transfer Mode: UDMA 5 (ATA-100)
>
>...which sounds encouraging. I think it's this high because I'm using
>the 80 conductor cables, and my WD EIDE drives support it, and, as I recall,
>I installed some Intel software update at one point in time (but NOT that
>infamous one, which trashed my system on another date (and there was one of
>those, shown as a recommended update, on the MS site, but they later removed
>it).
>
>So I'm wondering why yours is a bit lower above, and I expect it's due to
>some condition mentioned above not being met (or perhaps a really old BIOS)?

My drive supports ATA-66 but my old socket 7 motherboard is limited to
ATA-33.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Bill in Co.]

Franc Zabkar wrote:
> On Wed, 12 Mar 2008 21:53:31 -0600, "Bill in Co."
> <not_really_here@earthlink.net> put finger to keyboard and composed:
>
>> Franc Zabkar wrote:
>
>>> Everest Home Edition (for Windows) produces the following report for a
>>> Fujitsu MPF3204AT in a socket 7 box:
>>>
>>> Max. PIO Transfer Mode PIO 4
>>> Max. UDMA Transfer Mode UDMA 4 (ATA-66)
>>> Active UDMA Transfer Mode UDMA 2 (ATA-33)
>>>
>>> - Franc Zabkar
>>
>> Thanks, Frank. I'll look into both of them (and I already have an
>> older
>> freebie version of Everest (1.51)
>>
>> Hmmm. ...let's see...
>> OK, found it in Everest, under the "Storage, ATA" tab, I get:
>>
>> Max. PIO Transfer Mode: PIO 4
>> Max. UDMA Transfer Mode: UDMA 5 (ATA-100)
>> Active UDMA Transfer Mode: UDMA 5 (ATA-100)
>>
>> ...which sounds encouraging. I think it's this high because I'm using
>> the 80 conductor cables, and my WD EIDE drives support it, and, as I
>> recall,
>> I installed some Intel software update at one point in time (but NOT that
>> infamous one, which trashed my system on another date (and there was one
>> of
>> those, shown as a recommended update, on the MS site, but they later
>> removed
>> it).
>>
>> So I'm wondering why yours is a bit lower above, and I expect it's due to
>> some condition mentioned above not being met (or perhaps a really old
>> BIOS)?
>
> My drive supports ATA-66 but my old socket 7 motherboard is limited to
> ATA-33.

OK, but just out of curiosity, what is it about that motherboard,
specifically, that limits it - if you happen to know? (I mean technically,
from the hardware and/or software point of view). (You know, like the
example I gave of needing 80 conductor cables (using the 40 interleaving
ground wires) for higher data transfer rates to prevent the crosstalk
problem)

 

[Franc Zabkar]

On Thu, 13 Mar 2008 00:02:49 -0600, "Bill in Co."
<not_really_here@earthlink.net> put finger to keyboard and composed:

>Franc Zabkar wrote:

>> My drive supports ATA-66 but my old socket 7 motherboard is limited to
>> ATA-33.
>
>OK, but just out of curiosity, what is it about that motherboard,
>specifically, that limits it - if you happen to know? (I mean technically,
>from the hardware and/or software point of view). (You know, like the
>example I gave of needing 80 conductor cables (using the 40 interleaving
>ground wires) for higher data transfer rates to prevent the crosstalk
>problem)

AFAIK it's the chipset (SiS 5597/5598) that is the limitation.

The chipset datasheet states that its "Fast PCI IDE Master/Slave
Controller" supports ...

- IDE PIO Timing Mode 0, 1, 2 ,3 and 4
- Multiword DMA Mode 0, 1, 2
- Ultra DMA/33

See http://www.datasheetarchive.com/pdf/3078981.pdf

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[MEB]

"Bill in Co." <not_really_here@earthlink.net> wrote in message
news:u31sg$MhIHA.4684@TK2MSFTNGP06.phx.gbl...
| Franc Zabkar wrote:
| > On Wed, 12 Mar 2008 21:53:31 -0600, "Bill in Co."
| > <not_really_here@earthlink.net> put finger to keyboard and composed:
| >
| >> Franc Zabkar wrote:
| >
| >>> Everest Home Edition (for Windows) produces the following report for a
| >>> Fujitsu MPF3204AT in a socket 7 box:
| >>>
| >>> Max. PIO Transfer Mode PIO 4
| >>> Max. UDMA Transfer Mode UDMA 4 (ATA-66)
| >>> Active UDMA Transfer Mode UDMA 2 (ATA-33)
| >>>
| >>> - Franc Zabkar
| >>
| >> Thanks, Frank. I'll look into both of them (and I already have an
| >> older
| >> freebie version of Everest (1.51)
| >>
| >> Hmmm. ...let's see...
| >> OK, found it in Everest, under the "Storage, ATA" tab, I get:
| >>
| >> Max. PIO Transfer Mode: PIO 4
| >> Max. UDMA Transfer Mode: UDMA 5 (ATA-100)
| >> Active UDMA Transfer Mode: UDMA 5 (ATA-100)
| >>
| >> ...which sounds encouraging. I think it's this high because I'm
using
| >> the 80 conductor cables, and my WD EIDE drives support it, and, as I
| >> recall,
| >> I installed some Intel software update at one point in time (but NOT
that
| >> infamous one, which trashed my system on another date (and there was
one
| >> of
| >> those, shown as a recommended update, on the MS site, but they later
| >> removed
| >> it).
| >>
| >> So I'm wondering why yours is a bit lower above, and I expect it's due
to
| >> some condition mentioned above not being met (or perhaps a really old
| >> BIOS)?
| >
| > My drive supports ATA-66 but my old socket 7 motherboard is limited to
| > ATA-33.
|
| OK, but just out of curiosity, what is it about that motherboard,
| specifically, that limits it - if you happen to know? (I mean
technically,
| from the hardware and/or software point of view). (You know, like the
| example I gave of needing 80 conductor cables (using the 40 interleaving
| ground wires) for higher data transfer rates to prevent the crosstalk
| problem)
|
|

It revolves around the ATA/ATAPI standards that have been created and chips
which have this support HARD CODED into the actual adapter chip. The BIOS is
then created to support the chips supplied upon the mother board. If the
chip has no support for a newer standard {way of addressing/translation}
then nothing can be done to circumvent that limitation at the hard coded
level as the on-chip translation can not occur without that built-in
support. Flash upgrades for the BIOS or device may include code changes
which CAN be supported [chip-wise], but may not have been available or were
improperly coded when originally released.

Referencing:

http://www.t13.org/ - the master site
http://www.t13.org/Standards/Default.aspx?DocumentType=3
http://ata-atapi.com/
http://hddguru.com/content/en/documentation/2006.01.27-ATA-ATAPI-7/
http://hddguru.com/content/en/documentation/2006.01.27-ATA-ATAPI-5/
http://www.pcguide.com/ref/hdd/if/ide/stdATA4-c.html
http://www.storagereview.com/guide2000/ref/hdd/if/ide/stdATA4.html

The only way to consistently over-ride the limitations is to add an adapter
which provides the newer standards/translation, and a device which also
supports the target standard or newer.
Certain issues related to OS support can be addressed via modification to
drivers or additional software which provides that support, however, without
an adapter/chip set, which also supports the newer standard, these provide
nothing of real value.

--

MEB
http://peoplescounsel.orgfree.com
_________

 

[MEB]

One thing I should perhaps address related to flashing a motherboard BIOS,
is to remove/disconnect as many devices as possible, in addition to default
setting the BIOS/CMOS, to negate potential issue which may be involved when
the first reboot after the flash is done.
A previously recommended activity, was to only leave the floppy drive
connected and a 8bit or 16bit video card attached, though with the advent of
PCI video, that was changed as most had none of those cards, and VESA
standards were [by then] supported in the PCI/AGP video and BIOS.
For newer computers, without floppies and/or with in BIOS flash, these
issues have once again become at issue when motherboards are flashed. It
appears when flashing these boards, ANY error becomes difficult or nearly
impossible to recover from.

--

MEB
http://peoplescounsel.orgfree.com
_________

 

[Bill in Co.]

Franc Zabkar wrote:
> On Thu, 13 Mar 2008 00:02:49 -0600, "Bill in Co."
> <not_really_here@earthlink.net> put finger to keyboard and composed:
>
>> Franc Zabkar wrote:
>
>>> My drive supports ATA-66 but my old socket 7 motherboard is limited to
>>> ATA-33.
>>
>> OK, but just out of curiosity, what is it about that motherboard,
>> specifically, that limits it - if you happen to know? (I mean
>> technically,
>> from the hardware and/or software point of view). (You know, like the
>> example I gave of needing 80 conductor cables (using the 40 interleaving
>> ground wires) for higher data transfer rates to prevent the crosstalk
>> problem)
>
> AFAIK it's the chipset (SiS 5597/5598) that is the limitation.
>
> The chipset datasheet states that its "Fast PCI IDE Master/Slave
> Controller" supports ...
>
> - IDE PIO Timing Mode 0, 1, 2 ,3 and 4
> - Multiword DMA Mode 0, 1, 2
> - Ultra DMA/33
>
> See http://www.datasheetarchive.com/pdf/3078981.pdf

Wonder if there is aome updated driver available for that chipset?

Seems I did something like that for some Intel chips on my MB, at one time,
but I can't remember now. (Like maybe I initially didn't have Ultra DMA
available, and I had to update some drivers to get it (it's been too long to
recall).

 

[Franc Zabkar]

On Thu, 13 Mar 2008 20:51:55 -0600, "Bill in Co."
<not_really_here@earthlink.net> put finger to keyboard and composed:

>Franc Zabkar wrote:
>> On Thu, 13 Mar 2008 00:02:49 -0600, "Bill in Co."
>> <not_really_here@earthlink.net> put finger to keyboard and composed:
>>
>>> Franc Zabkar wrote:
>>
>>>> My drive supports ATA-66 but my old socket 7 motherboard is limited to
>>>> ATA-33.
>>>
>>> OK, but just out of curiosity, what is it about that motherboard,
>>> specifically, that limits it - if you happen to know? (I mean
>>> technically,
>>> from the hardware and/or software point of view). (You know, like the
>>> example I gave of needing 80 conductor cables (using the 40 interleaving
>>> ground wires) for higher data transfer rates to prevent the crosstalk
>>> problem)
>>
>> AFAIK it's the chipset (SiS 5597/5598) that is the limitation.
>>
>> The chipset datasheet states that its "Fast PCI IDE Master/Slave
>> Controller" supports ...
>>
>> - IDE PIO Timing Mode 0, 1, 2 ,3 and 4
>> - Multiword DMA Mode 0, 1, 2
>> - Ultra DMA/33
>>
>> See http://www.datasheetarchive.com/pdf/3078981.pdf
>
>Wonder if there is aome updated driver available for that chipset?
>
>Seems I did something like that for some Intel chips on my MB, at one time,
>but I can't remember now. (Like maybe I initially didn't have Ultra DMA
>available, and I had to update some drivers to get it (it's been too long to
>recall).

It's not a driver limitation. It's an electronic limitation of the
chipset.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Bill in Co.]

Franc Zabkar wrote:
> On Thu, 13 Mar 2008 20:51:55 -0600, "Bill in Co."
> <not_really_here@earthlink.net> put finger to keyboard and composed:
>
>> Franc Zabkar wrote:
>>> On Thu, 13 Mar 2008 00:02:49 -0600, "Bill in Co."
>>> <not_really_here@earthlink.net> put finger to keyboard and composed:
>>>
>>>> Franc Zabkar wrote:
>>>
>>>>> My drive supports ATA-66 but my old socket 7 motherboard is limited to
>>>>> ATA-33.
>>>>
>>>> OK, but just out of curiosity, what is it about that motherboard,
>>>> specifically, that limits it - if you happen to know? (I mean
>>>> technically,
>>>> from the hardware and/or software point of view). (You know, like
>>>> the
>>>> example I gave of needing 80 conductor cables (using the 40
>>>> interleaving
>>>> ground wires) for higher data transfer rates to prevent the crosstalk
>>>> problem)
>>>
>>> AFAIK it's the chipset (SiS 5597/5598) that is the limitation.
>>>
>>> The chipset datasheet states that its "Fast PCI IDE Master/Slave
>>> Controller" supports ...
>>>
>>> - IDE PIO Timing Mode 0, 1, 2 ,3 and 4
>>> - Multiword DMA Mode 0, 1, 2
>>> - Ultra DMA/33
>>>
>>> See http://www.datasheetarchive.com/pdf/3078981.pdf
>>
>> Wonder if there is aome updated driver available for that chipset?
>>
>> Seems I did something like that for some Intel chips on my MB, at one
>> time,
>> but I can't remember now. (Like maybe I initially didn't have Ultra
>> DMA
>> available, and I had to update some drivers to get it (it's been too long
>> to
>> recall).
>
> It's not a driver limitation. It's an electronic limitation of the
> chipset.

OK. At least in my case, I had thought I had updated some Intel driver(s)
that enabled some additional features or capabilities of some chipset I had
on my Motherboard.

But maybe that is impossible, and I'm misrembering it. So let me ask this
question then, since I don't fully understand it. Isn't it possible that
some chipsets CAN have new software drivers that can enable additional
features, provided the manufacturer of the chipset allowed for such
upgrading?

But your chipset may be older than mine, too. (I have a Dell 4100, circa
2000, as I recall)

 

[Franc Zabkar]

On Fri, 14 Mar 2008 00:28:06 -0600, "Bill in Co."
<not_really_here@earthlink.net> put finger to keyboard and composed:

>I had thought I had updated some Intel driver(s)
>that enabled some additional features or capabilities of some chipset I had
>on my Motherboard.
>
>But maybe that is impossible, and I'm misrembering it. So let me ask this
>question then, since I don't fully understand it. Isn't it possible that
>some chipsets CAN have new software drivers that can enable additional
>features, provided the manufacturer of the chipset allowed for such
>upgrading?

Yes it's possible, and it's also possible to enable additional chipset
features via the BIOS. In fact I've tweaked various chipset registers
from within Windows using WPCREDIT and WPCRSET. IIRC this gave me a
10% performance increase in my RAM benchmarks.

See ...

Chipset BIOS tweaking the M571:
http://www.m571.com/m571/tweek.htm

>But your chipset may be older than mine, too. (I have a Dell 4100, circa
>2000, as I recall)

Mine dates back to around 1998.

This URL seems to itemise your chipset components:
http://gentoo-wiki.com/HARDWARE_Gentoo_on_Dell_Inspiron_4100

If you feel like tweaking or just viewing your chipset registers, then
I suggest you find the datasheets for Intel's 82830 and 82801 north
and southbridges and then experiment with WPCREDIT.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

 

[Bill in Co.]

Franc Zabkar wrote:
> On Fri, 14 Mar 2008 00:28:06 -0600, "Bill in Co."
> <not_really_here@earthlink.net> put finger to keyboard and composed:
>
>> I had thought I had updated some Intel driver(s)
>> that enabled some additional features or capabilities of some chipset I
>> had
>> on my Motherboard.
>>
>> But maybe that is impossible, and I'm misrembering it. So let me ask
>> this
>> question then, since I don't fully understand it. Isn't it possible
>> that
>> some chipsets CAN have new software drivers that can enable additional
>> features, provided the manufacturer of the chipset allowed for such
>> upgrading?
>
> Yes it's possible, and it's also possible to enable additional chipset
> features via the BIOS. In fact I've tweaked various chipset registers
> from within Windows using WPCREDIT and WPCRSET. IIRC this gave me a
> 10% performance increase in my RAM benchmarks.
>
> See ...
>
> Chipset BIOS tweaking the M571:
> http://www.m571.com/m571/tweek.htm
>
>> But your chipset may be older than mine, too. (I have a Dell 4100, circa
>> 2000, as I recall)
>
> Mine dates back to around 1998.
>
> This URL seems to itemise your chipset components:
> http://gentoo-wiki.com/HARDWARE_Gentoo_on_Dell_Inspiron_4100
>
> If you feel like tweaking or just viewing your chipset registers, then
> I suggest you find the datasheets for Intel's 82830 and 82801 north
> and southbridges and then experiment with WPCREDIT.
>
> - Franc Zabkar

OK, thanks for the info, Frank.